A PQC-native Layer 1 blockchain.

Keypair, faucet, transfer, and basic contract lifecycle are available for experimentation on the public testnet. Staking, governance, and some advanced or operator paths are not yet production-complete.

Dytallix implements NIST-standardized PQC primitives at every layer of the protocol. ML-DSA-65 for all transaction signing and validator votes. ML-KEM-768 for all P2P transport. Canonical Bech32m addresses. No ECDSA. No hybrid mode. No legacy accounts. Every address, every signature, every handshake is quantum-secure from inception.

Dytallix is a small startup developing a post-quantum product ecosystem.

Co-founders: Andrew Holland and Rick Glenn. About Dytallix.

Discovery tags: post-quantum-cryptography · pqc · quantum-safe · layer1 · blockchain · testnet · ml-dsa · ml-kem · bech32m · rust · webassembly · wasm · sdk · cli · faucet · block-explorer

• Publish public testnet developer path: keypair, faucet, transfer, and basic contract lifecycle.
• Open-source SDK, CLI, PQC primitives, node source, faucet source, explorer surface map, docs, and protocol contracts.
• Document NIST PQC baseline with ML-DSA-65 signing, ML-KEM-768 transport, and Bech32m addresses.
• Complete staking, governance, and advanced operator paths for production readiness.
• Formalize VRF construction, handshake analysis, and side-channel review before mainnet.
• Expand third-party audits, protocol review, and contributor onboarding.

Adversaries are capturing encrypted traffic today to decrypt it once a cryptographically relevant quantum computer arrives — the "Harvest Now, Decrypt Later" threat is not future risk, it is present risk. Any blockchain that still accepts ECDSA signatures is accumulating cryptographic debt that a CRQC will collect on.

Dytallix is built so you never have to migrate.

Install Rust with rustup. That provides the Rust toolchain, cargo, and target management used throughout the Dytallix Rust repositories.

If you plan to build WASM contracts locally, add the standard target:

rustup target add wasm32-unknown-unknown

These are the three developer milestones Dytallix is optimizing for:

1. First keypair: under 60 seconds

   Generate your first ML-DSA-65 keypair and print a D-Addr:

   git clone https://github.com/DytallixHQ/dytallix-sdk
   cd dytallix-sdk
   cargo run -p dytallix-sdk --example first-keypair

   Start here: first-keypair example

2. First transaction on testnet: 2-3 minutes

   Create a funded sender wallet, create a separate recipient wallet, then submit and verify a real transaction:

   cargo install --git https://github.com/DytallixHQ/dytallix-sdk.git dytallix-cli --bin dytallix
   dytallix init
   dytallix wallet create --name recipient
   dytallix wallet list
   dytallix wallet switch default
   dytallix send <recipient-daddr> 100
   dytallix wallet switch recipient
   dytallix balance

   Use a different recipient address than the one printed by dytallix init (do not self-send).

   Continue with: first-transaction example · Explorer · Releases

3. First contract build: under 15 minutes

   Build a minimal WASM contract now. The default public gateway accepts POST /contracts/deploy and POST /contracts/call; use a direct node endpoint or a local node only when you want local testing or custom infrastructure.

   rustup target add wasm32-unknown-unknown
   cargo build --manifest-path examples/contracts/minimal_contract/Cargo.toml --target wasm32-unknown-unknown --release

   Continue with: deploy-contract example · dytallix-contracts · Docs

→ SDK repo · Releases · Full getting started guide

Canonical public integration guidance lives in dytallix-docs. The live website frontend and hosted explorer UI remain public runtime surfaces at dytallix.com, and the live faucet backend source is public in dytallix-faucet.

The dytallix-explorer repository remains a docs-only surface map for the explorer endpoints. The node repo now has clean-checkout production provenance from the public deployment.

All parameter sizes are exact. ML-DSA-65 public key is 1,952 bytes. Signature is 3,309 bytes. A single mistyped character in any address fails checksum validation. There is no classical fallback.

Reference implementation and developer docs: dytallix-pqc · PQC docs · CLI tools

We publish the parts of this protocol we are not fully satisfied with. A project that knows its own edges is more trustworthy than one that claims to have none.

• VRF construction — the current leader election uses a deterministic-signature-derived VRF that has not been formally proven to satisfy full VRF security properties. It is a practical stand-in pending a standardized PQC VRF.
• Isochronous rejection sampling — ML-DSA signing uses a variable-iteration rejection sampling loop. The implementation is designed to be constant-time but a formal side-channel analysis has not been completed.
• Signature aggregation — ML-DSA does not support constant-size non-interactive aggregation like BLS. Checkpoint finality uses Bitfield-Compressed Merkle Proofs instead. If efficient PQC aggregation is standardized we will adopt it.
• Handshake formal verification — the Noise_IK_PQC handshake has not been formally analyzed with ProVerif or Tamarin. Scheduled for pre-mainnet audit.

Full technical detail and current long-form materials live at Resources.

Long-form protocol papers and supporting materials are indexed on the main resources page:

• Resources — whitepapers, protocol background, and supporting materials

We welcome contributors to the Dytallix protocol and developer tools.

We are looking for cryptographers to audit the PQC implementation, systems engineers to stress-test the testnet, and developers to build the first PQC-native dApps.

Start here: dytallix-sdk

If you find something broken or something that could be better, open an issue. If you want to talk first, join Discord.

Website · Documentation · Resources · Discord · Explorer · Faucet · X

Testnet is live. Mainnet is on the roadmap. Core protocol and developer tooling are open source; the public explorer and faucet repositories document the live service surfaces.